A deterministic matching method for exact matchings to compare the outcome of different interventions

TL;DR

This paper introduces a deterministic exact matching method for observational data that ensures low bias, interpretability, and provably unique results, improving upon existing algorithms in social and health sciences research.

Contribution

It develops a novel deterministic matching approach that uses all available information for exact matches, guaranteeing unique, bias-free results and compatibility with other methods.

Findings

Method guarantees low bias in matching results

Results are provably unique and deterministic

Efficient computation demonstrated on survey data

Abstract

Statistical matching methods are widely used in the social and health sciences to estimate causal effects using observational data. Often the objective is to find comparable groups with similar covariate distributions in a dataset, with the aim to reduce bias in a random experiment. We aim to develop a foundation for deterministic methods which provide results with low bias, while retaining interpretability. The proposed method matches on the covariates and calculates all possible maximal exact matchesfor a given dataset without adding numerical errors. Notable advantages of our method over existing matching algorithms are that all available information for exact matches is used, no additional bias is introduced, it can be combined with other matching methods for inexact matching to reduce pruning and that the result is calculated in a fast and deterministic way. For a given dataset the result is therefore provably unique for exact matches in the mathematical sense. We provide proofs, instructions for implementation as well as a numerical example calculated for comparison on a complete survey.